modelmix 5.0.6 → 5.1.2

package/README.md

@@ -15,9 +15,9 @@ Ever found yourself wanting to integrate AI models into your projects but worrie
 - [Token Usage Tracking](#-token-usage-tracking)
 - [Prompt Caching](#-prompt-caching)
 - [Model Context Protocol (MCP) Integration](#-model-context-protocol-mcp-integration)
-- [Enabling Debug Mode](#-enabling-debug-mode)
-- [Bottleneck Integration](#-bottleneck-integration)
 - [Retry (Opt-In)](#-retry-optin)
+- [Bottleneck Integration](#-bottleneck-integration)
+- [Enabling Debug Mode](#-enabling-debug-mode)
 - [Instance Plugins](#-instance-plugins)
 - [ModelMix Class Overview](#-modelmix-class-overview)
 - [Contributing](#-contributing)
@@ -117,9 +117,9 @@ const ETH = ModelMix.new()
 console.log(ETH.price);
 ```
 
-**This example uses providers with free quotas (OpenRouter, Groq, Cerebras) - just get the API key and you're ready to go. If one model runs out of quota, ModelMix automatically falls back to the next model in the chain.**
+**This example uses providers with free quotas (Groq, Cerebras, and Together). OpenRouter is disabled because its GPT-OSS 120B route is no longer free. If one model runs out of quota, ModelMix automatically falls back to the next model in the chain.**
 ```javascript
-ModelMix.new()
+ModelMix.new({ mix: { openrouter: false } })
   .gptOss()
   .kimiK25()
   .hermes3()
@@ -136,63 +136,59 @@ This pattern allows you to:
 ## ⚡️ Shorthand Methods
 
 ModelMix provides convenient shorthand methods for quickly accessing different AI models.
-Here's a comprehensive list of available methods:
-
-| Method              | Provider   | Model                        | Price (I/O) per 1 M tokens |
-| ------------------- | ---------- | ---------------------------- | -------------------------- |
-| `gpt56sol()`        | OpenAI     | gpt-5.6-sol                  | [\$5.00/\$30.00][1]      |
-| `gpt56terra()`      | OpenAI     | gpt-5.6-terra                | [\$2.00/\$12.00][1]      |
-| `gpt56luna()`       | OpenAI     | gpt-5.6-luna                 | [\$0.20/\$1.20][1]       |
-| `gpt55()`           | OpenAI     | gpt-5.5                      | [\$5.00/\$30.00][1]      |
-| `gpt54()`           | OpenAI     | gpt-5.4                      | [\$2.50/\$15.00][1]      |
-| `gpt54mini()`       | OpenAI     | gpt-5.4-mini                 | [\$0.75/\$4.50][1]       |
-| `gpt54nano()`       | OpenAI     | gpt-5.4-nano                 | [\$0.20/\$1.25][1]       |
-| `gpt53codex()`      | OpenAI     | gpt-5.3-codex                | [\$1.25/\$14.00][1]      |
-| `gpt52()`           | OpenAI     | gpt-5.2                      | [\$1.75/\$14.00][1]      |
-| `gpt51()`           | OpenAI     | gpt-5.1                      | [\$1.25/\$10.00][1]      |
-| `gpt5mini()`        | OpenAI     | gpt-5-mini                   | [\$0.25/\$2.00][1]       |
-| `gpt5nano()`        | OpenAI     | gpt-5-nano                   | [\$0.05/\$0.40][1]       |
-| `gpt41()`           | OpenAI     | gpt-4.1                      | [\$2.00/\$8.00][1]       |
-| `gpt41mini()`       | OpenAI     | gpt-4.1-mini                 | [\$0.40/\$1.60][1]       |
-| `gpt41nano()`       | OpenAI     | gpt-4.1-nano                 | [\$0.10/\$0.40][1]       |
-| `gptOss()`          | Together   | gpt-oss-120B                 | [\$0.15/\$0.60][7]       |
-| `fable5()`          | Anthropic  | claude-fable-5               | [\$10.00/\$50.00][2]     |
-| `opus5()`           | Anthropic  | claude-opus-5                | [\$5.00/\$25.00][2]      |
-| `opus48()`          | Anthropic  | claude-opus-4-8              | [\$5.00/\$25.00][2]      |
-| `opus47()`          | Anthropic  | claude-opus-4-7              | [\$5.00/\$25.00][2]      |
-| `opus46()`          | Anthropic  | claude-opus-4-6              | [\$5.00/\$25.00][2]      |
-| `sonnet5()`         | Anthropic  | claude-sonnet-5              | [\$3.00/\$15.00][2]      |
-| `sonnet46()`        | Anthropic  | claude-sonnet-4-6            | [\$3.00/\$15.00][2]      |
-| `haiku45()`         | Anthropic  | claude-haiku-4-5-20251001    | [\$1.00/\$5.00][2]       |
-| `gemini31pro()`     | Google     | gemini-3.1-pro-preview       | [\$2.00/\$12.00][3]      |
-| `gemini37flash()`   | Google     | gemini-3.7-flash             | [\$0.75/\$3.75][3]       |
-| `gemini36flash()`   | Google     | gemini-3.6-flash             | [\$0.75/\$3.75][3]       |
-| `gemini35flash()`    | Google     | gemini-3.5-flash              | [\$0.75/\$4.50][3]       |
-| `gemini35flashLite()`| Google     | gemini-3.5-flash-lite          | [\$0.30/\$2.50][3]       |
-| `gemini31flashLite()`| Google     | gemini-3.1-flash-lite-preview | [\$0.25/\$1.50][3]       |
-| `grok46()`          | Grok       | grok-4.6                     | [\$2.00/\$6.00][6]       |
-| `grok45()`          | Grok       | grok-4.5                     | [\$2.00/\$6.00][6]       |
-| `grok43()`          | Grok       | grok-4.3                     | [\$1.25/\$2.50][6]       |
-| `grok420multiAgent()`| Grok      | grok-4.20-multi-agent-0309   | [\$1.25/\$2.50][6]       |
-| `grok420()`         | Grok       | grok-4.20-0309 (†)           | [\$1.25/\$2.50][6]       |
-| `qwen35397b()`      | OpenRouter | qwen/qwen3.5-397b-a17b       | [\$0.385/\$2.45][14]     |
-| `qwen36plus()`      | Fireworks  | qwen3p6-plus                 | [\$0.50/\$3.00][10]      |
-| `qwen37plus()`      | Fireworks  | models/qwen3p7-plus          | [\$0.40/\$1.60][10]      |
-| `qwen38max()`       | Fireworks  | qwen3p8-2p4t-a95b            | [\$2.00/\$6.00][10]      |
-| `deepseekV4Flash()` | Fireworks  | models/deepseek-v4-flash      | [\$0.14/\$0.28][10]      |
-| `deepseekV4Pro()`   | Fireworks  | models/deepseek-v4-pro-0813  | [\$1.32/\$3.96][12]      |
-| `GLM52()`           | Together   | zai-org/GLM-5.2              | [\$1.40/\$4.40][7]       |
-| `GLM51()`           | Fireworks  | models/glm-5p1               | [\$1.05/\$3.50][10]      |
-| `minimaxM3()`       | MiniMax    | MiniMax-M3                   | [\$0.30/\$1.20][9]       |
-| `minimaxM27()`      | MiniMax    | MiniMax-M2.7                 | [\$0.30/\$1.20][9]       |
-| `sonar()`           | Perplexity | sonar                        | [\$1.00/\$1.00][4]       |
-| `sonarPro()`        | Perplexity | sonar-pro                    | [\$3.00/\$15.00][4]      |
-| `hermes470b()`      | OpenRouter | nousresearch/hermes-4-70b    | [\$0.13/\$0.40][13]      |
-| `hermes4405b()`     | OpenRouter | nousresearch/hermes-4-405b   | [\$1.00/\$3.00][13]      |
-| `hermes3()`         | Lambda     | Hermes-3-Llama-3.1-405B-FP8  | [\$0.80/\$0.80][8]       |
-| `kimiK3()`          | Moonshot   | kimi-k3                      | [\$3.00/\$15.00][11]     |
-| `kimiK25()`         | Together   | Kimi-K2.5                    | [\$0.50/\$2.80][7]       |
-| `kimiK26()`         | Fireworks  | models/kimi-k2p6             | [\$0.95/\$4.00][10]      |
+
+| Method | Provider | Model | Input / 1M | Output / 1M |
+| --- | --- | --- | ---: | ---: |
+| `gpt56sol()` | OpenAI | gpt-5.6-sol | [\$5.00][1] | [\$30.00][1] |
+| `gpt56terra()` | OpenAI | gpt-5.6-terra | [\$2.00][1] | [\$12.00][1] |
+| `gpt56luna()` | OpenAI | gpt-5.6-luna | [\$0.20][1] | [\$1.20][1] |
+| `gpt55()` | OpenAI | gpt-5.5 | [\$5.00][1] | [\$30.00][1] |
+| `gpt54()` | OpenAI | gpt-5.4 | [\$2.50][1] | [\$15.00][1] |
+| `gpt54mini()` | OpenAI | gpt-5.4-mini | [\$0.75][1] | [\$4.50][1] |
+| `gpt54nano()` | OpenAI | gpt-5.4-nano | [\$0.20][1] | [\$1.25][1] |
+| `gpt53codex()` | OpenAI | gpt-5.3-codex | [\$1.25][1] | [\$14.00][1] |
+| `gpt52()` | OpenAI | gpt-5.2 | [\$1.75][1] | [\$14.00][1] |
+| `gpt51()` | OpenAI | gpt-5.1 | [\$1.25][1] | [\$10.00][1] |
+| `gpt5mini()` | OpenAI | gpt-5-mini | [\$0.25][1] | [\$2.00][1] |
+| `gpt5nano()` | OpenAI | gpt-5-nano | [\$0.05][1] | [\$0.40][1] |
+| `gptOss()` | Together | gpt-oss-120B | [\$0.15][7] | [\$0.60][7] |
+| `fable5()` | Anthropic | claude-fable-5 | [\$10.00][2] | [\$50.00][2] |
+| `opus5()` | Anthropic | claude-opus-5 | [\$5.00][2] | [\$25.00][2] |
+| `opus48()` | Anthropic | claude-opus-4-8 | [\$5.00][2] | [\$25.00][2] |
+| `opus47()` | Anthropic | claude-opus-4-7 | [\$5.00][2] | [\$25.00][2] |
+| `opus46()` | Anthropic | claude-opus-4-6 | [\$5.00][2] | [\$25.00][2] |
+| `sonnet5()` | Anthropic | claude-sonnet-5 | [\$3.00][2] | [\$15.00][2] |
+| `sonnet46()` | Anthropic | claude-sonnet-4-6 | [\$3.00][2] | [\$15.00][2] |
+| `haiku45()` | Anthropic | claude-haiku-4-5-20251001 | [\$1.00][2] | [\$5.00][2] |
+| `gemini31pro()` | Google | gemini-3.1-pro-preview | [\$2.00][3] | [\$12.00][3] |
+| `gemini37flash()` | Google | gemini-3.7-flash | [\$0.75][3] | [\$3.75][3] |
+| `gemini36flash()` | Google | gemini-3.6-flash | [\$0.75][3] | [\$3.75][3] |
+| `gemini35flash()` | Google | gemini-3.5-flash | [\$0.75][3] | [\$4.50][3] |
+| `gemini35flashLite()` | Google | gemini-3.5-flash-lite | [\$0.30][3] | [\$2.50][3] |
+| `gemini31flashLite()` | Google | gemini-3.1-flash-lite-preview | [\$0.25][3] | [\$1.50][3] |
+| `grok46()` | Grok | grok-4.6 | [\$2.00][6] | [\$6.00][6] |
+| `grok45()` | Grok | grok-4.5 | [\$2.00][6] | [\$6.00][6] |
+| `grok43()` | Grok | grok-4.3 | [\$1.25][6] | [\$2.50][6] |
+| `grok420multiAgent()` | Grok | grok-4.20-multi-agent-0309 | [\$1.25][6] | [\$2.50][6] |
+| `grok420()` | Grok | grok-4.20-0309 (†) | [\$1.25][6] | [\$2.50][6] |
+| `qwen35397b()` | OpenRouter | qwen/qwen3.5-397b-a17b | [\$0.385][14] | [\$2.45][14] |
+| `qwen36plus()` | Fireworks | qwen3p6-plus | [\$0.50][10] | [\$3.00][10] |
+| `qwen37plus()` | Fireworks | models/qwen3p7-plus | [\$0.40][10] | [\$1.60][10] |
+| `qwen38max()` | Fireworks | qwen3p8-2p4t-a95b | [\$2.00][10] | [\$6.00][10] |
+| `deepseekV4Flash()` | Fireworks | models/deepseek-v4-flash | [\$0.14][10] | [\$0.28][10] |
+| `deepseekV4Pro()` | Fireworks | models/deepseek-v4-pro-0813 | [\$1.32][12] | [\$3.96][12] |
+| `GLM52()` | Together | zai-org/GLM-5.2 | [\$1.40][7] | [\$4.40][7] |
+| `GLM51()` | Fireworks | models/glm-5p1 | [\$1.05][10] | [\$3.50][10] |
+| `minimaxM3()` | MiniMax | MiniMax-M3 | [\$0.30][9] | [\$1.20][9] |
+| `minimaxM27()` | MiniMax | MiniMax-M2.7 | [\$0.30][9] | [\$1.20][9] |
+| `sonar()` | Perplexity | sonar | [\$1.00][4] | [\$1.00][4] |
+| `sonarPro()` | Perplexity | sonar-pro | [\$3.00][4] | [\$15.00][4] |
+| `hermes470b()` | OpenRouter | nousresearch/hermes-4-70b | [\$0.13][13] | [\$0.40][13] |
+| `hermes4405b()` | OpenRouter | nousresearch/hermes-4-405b | [\$1.00][13] | [\$3.00][13] |
+| `hermes3()` | Lambda | Hermes-3-Llama-3.1-405B-FP8 | [\$0.80][8] | [\$0.80][8] |
+| `kimiK3()` | Moonshot | kimi-k3 | [\$3.00][11] | [\$15.00][11] |
+| `kimiK25()` | Together | Kimi-K2.5 | [\$0.50][7] | [\$2.80][7] |
+| `kimiK26()` | Fireworks | models/kimi-k2p6 | [\$0.95][10] | [\$4.00][10] |
 
 Gemini 3.7 Flash and 3.6 Flash use Google's introductory standard pricing through December 31, 2026; standard rates double on January 1, 2027.
 
@@ -249,6 +245,7 @@ ModelMix.new().effort(-1).minimaxM3().addText('...').message();
 ### Provider-specific behavior
 
 - **Gemini:** Gemini 3+ uses bands 0–24 / 25–49 / 50–74 / 75–100. Gemini 3.7 Flash clamps these bands to `low` / `low` / `medium` / `high`; `-1` leaves its native `medium` default unchanged. Gemini 2.5 maps 0–100 to `thinkingBudget`.
+- **GPT-5.6:** `100` maps to `max`; 80–99 remains `xhigh`.
 - **DeepSeek:** `↑` means thinking is enabled; `off` means it is disabled.
 - **MiniMax:** `off` maps to `thinking.disabled`; `adaptive` maps to `thinking.type=adaptive`.
 - **Anthropic:** Claude 5, Fable, Opus 4.6+, and Sonnet 4.6+ use adaptive thinking with `output_config.effort`. Sonnet 4.5 and Haiku 4.5 use `thinking.type=enabled` with `budget_tokens`.
@@ -841,56 +838,6 @@ This simple integration allows your model to:
 
 The Model Context Protocol makes it easy to add any capability to your models, from web search to code execution, database queries, or custom functions. All with just a few lines of code!
 
-## 🐛 Enabling Debug Mode
-
-To activate debug mode in ModelMix and view detailed request information, follow these two steps:
-
-1. In the ModelMix constructor, include a `debug` level in the configuration:
-
-   ```javascript
-   const mix = ModelMix.new({
-     config: {
-       debug: 4 // 0=silent, 1=minimal, 2=summary, 3=full (no truncate), 4=verbose (raw details)
-       // ... other configuration options ...
-     }
-   });
-   ```
-
-2. When running your script from the command line, use the `DEBUG=ModelMix*` prefix:
-
-   ```
-   DEBUG=ModelMix* node your_script.js
-   ```
-
-When you run your script this way, you'll see detailed information about the requests in the console, including the configuration and options used for each AI model request.
-
-This information is valuable for debugging and understanding how ModelMix is processing your requests.
-
-## 🚦 Bottleneck Integration
-
-ModelMix now uses Bottleneck for efficient rate limiting of API requests. This integration helps prevent exceeding API rate limits and ensures smooth operation when working with multiple models or high request volumes.
-
-### How it works:
-
-1. **Configuration**: Bottleneck is configured in the ModelMix constructor. You can customize the settings or use the default configuration:
-
-```javascript
-const setup = {
-    config: {
-        bottleneck: {
-            maxConcurrent: 8,     // Maximum number of concurrent requests
-            minTime: 500          // Minimum time between requests (in ms)
-        }
-    }
-};
-```
-
-2. **Rate Limiting**: When you make a request using any of the attached models, Bottleneck automatically manages the request flow based on the configured settings.
-
-3. **Automatic Queueing**: If the rate limit is reached, Bottleneck will automatically queue subsequent requests and process them as capacity becomes available.
-
-This integration ensures that your application respects API rate limits while maximizing throughput, providing a robust solution for managing multiple AI model interactions.
-
 ## 🔁 Retry (Opt-In)
 
 ModelMix supports optional intra-model retries for transient HTTP failures. When enabled, it retries the same provider before moving to fallback models.
@@ -914,6 +861,35 @@ Behavior summary:
 - If retry is enabled, ModelMix retries the same model only for configured transient status codes.
 - After retries are exhausted (or for non-retryable errors), ModelMix continues with normal fallback chain.
 
+## 🚦 Bottleneck Integration
+
+ModelMix uses Bottleneck for efficient rate limiting of API requests.
+
+```javascript
+const setup = {
+    config: {
+        bottleneck: {
+            maxConcurrent: 8,
+            minTime: 500
+        }
+    }
+};
+```
+
+Attached models share this limiter, which queues requests when capacity is exhausted.
+
+## 🐛 Enabling Debug Mode
+
+Set `config.debug` to `0` (silent), `1` (minimal), `2` (summary), `3` (full), or `4` (verbose raw details), then run with `DEBUG=ModelMix*`:
+
+```javascript
+const mix = ModelMix.new({ config: { debug: 4 } });
+```
+
+```bash
+DEBUG=ModelMix* node your-script.js
+```
+
 ## 🔌 Instance Plugins
 
 Plugins wrap one ModelMix instance without changing global behavior. They run in registration order after templates are rendered and before provider-specific request conversion:
@@ -958,7 +934,7 @@ The separately publishable `@modelmix/rlm` workspace package keeps document pars
 const { ModelMix } = require('modelmix');
 const { rlm } = require('@modelmix/rlm');
 
-const fast = ModelMix.new().gpt41mini();
+const fast = ModelMix.new().gpt5nano();
 
 const result = await ModelMix.new()
     .gpt56luna()

package/RLM_PLUGIN_SPEC.md

@@ -204,14 +204,14 @@ Proposed configuration:
 ```js
 const workers = {
     fast: {
-        model: ModelMix.new().gpt41nano(),
+        model: ModelMix.new().gpt5nano(),
         intelligence: 1,
         cost: 1,
         speed: 5,
         description: 'Extraction, classification, and simple transformations'
     },
     balanced: {
-        model: ModelMix.new().gpt41mini(),
+        model: ModelMix.new().gpt5nano(),
         intelligence: 3,
         cost: 2,
         speed: 4,

package/demo/fallback.js

@@ -15,7 +15,7 @@ const mmix = new ModelMix({
     }
 });
 
-mmix.sonnet46({ config: { url: 'fail' } }).gpt41nano();
+mmix.sonnet46({ config: { url: 'fail' } }).gpt5nano();
 
 async function main() {
     mmix.addText('hola, como estas?');
@@ -23,4 +23,4 @@ async function main() {
     console.log(response);
 }
 
-main();
+main();

package/demo/free.js

@@ -1,13 +1,12 @@
 import { ModelMix } from '../index.js';
 try { process.loadEnvFile(); } catch {}
 
-const ai = ModelMix.new({ config: { debug: 2 } })
+const ai = ModelMix.new({ config: { debug: 2 }, mix: { openrouter: false } })
     .gptOss()
     .kimiK25()
     .hermes3()
     .addText('What is the capital of France?');
 
 const response = await ai.message();
-console.log('Response from Claude via OpenRouter:', response);
-
+console.log('Response from a free-tier provider:', response);
 

package/demo/gemini.js

@@ -20,8 +20,8 @@ const flash = await mmix.gemini37flash()
 
 console.log(flash);
 
-// Using gemini3pro (Gemini 3 Pro) with custom config
-console.log("\n" + '--------| gemini3pro() with JSON response |--------');
+// Using Gemini 3.1 Pro with custom config
+console.log("\n" + '--------| gemini31pro() with JSON response |--------');
 const pro = mmix.new().gemini31pro();
 
 pro.addText('Give me a fun fact about cats');

package/demo/json.js

@@ -6,7 +6,7 @@ const model = await ModelMix.new({ options: { max_tokens: 10000 }, config: { deb
     // .gptOss()
     // .o4mini()
     // .gpt45()
-    // .gemini25flash()
+    // .gemini37flash()
     .addText("Name and capital of 3 South American countries.")
 
 const jsonResult = await model.json([{
@@ -24,4 +24,4 @@ const jsonResult = await model.json([{
 }], { addNote: true });
 
 console.log(jsonResult);
-console.log(model.lastRaw.tokens);
+console.log(model.lastRaw.tokens);

package/demo/mcp-simple.js

@@ -8,7 +8,7 @@ async function simpleCalculator() {
     console.log('\n=== Smart Calculator ===');
 
     const mmix = ModelMix.new()
-        .gpt41nano()
+        .gpt5nano()
         .setSystem('You are a smart calculator. Use the available tools to perform calculations.');
 
     // Add custom tool for advanced operations
@@ -93,7 +93,7 @@ async function contentGenerator() {
     console.log('\n=== Content Generator ===');
 
     const mmix = ModelMix.new({ config: { debug: 2, max_history: 1 } })
-        .gemini3flash()
+        .gemini37flash()
         .setSystem('You are a creative assistant that can generate different types of content.');
 
     // Tool for generating passwords

package/demo/mcp-tools.js

@@ -9,7 +9,7 @@ async function example1() {
     console.log('\n=== Example 1: Common Tools ===');
     
     const mmix = ModelMix.new({ config: { max_history: 10 } })
-        .gpt41nano();
+        .gpt5nano();
     
     // Add custom tools
     mmix.addTool({
@@ -66,7 +66,7 @@ async function example2() {
     console.log('\n=== Example 2: Custom Tool - Read Files ===');
     
     const mmix = ModelMix.new({ config: { max_history: 10 } })
-        .gpt41nano();
+        .gpt5nano();
 
     // Register custom tool for reading files
     mmix.addTool({
@@ -102,7 +102,7 @@ async function example3() {
     console.log('\n=== Example 3: HTTP Request Tool ===');
     
     const mmix = ModelMix.new({ config: { max_history: 10 } })
-        .gpt41nano();
+        .gpt5nano();
 
     // Register tool for making HTTP requests
     mmix.addTool({
@@ -158,7 +158,7 @@ async function example4() {
     console.log('\n=== Example 4: Multiple Tools Working Together ===');
     
     const mmix = ModelMix.new({ config: { max_history: 10 } })
-        .gpt41nano();
+        .gpt5nano();
 
     // Register multiple tools at once
     mmix.addTools([
@@ -248,7 +248,7 @@ async function example5() {
     console.log('\n=== Example 5: External MCP + Local Tools ===');
     
     const mmix = ModelMix.new({ config: { max_history: 10 } })
-        .gpt41nano();
+        .gpt5nano();
 
     // Add external MCP (if available)
     try {
@@ -341,7 +341,7 @@ async function runExamples() {
         console.log('\n✅ All examples completed');
         
         // Show registered tools
-        const mmix = ModelMix.new().gpt41nano();
+        const mmix = ModelMix.new().gpt5nano();
         mmix.addTool({
             name: "example_tool",
             description: "Example tool for demonstration",

package/demo/mcp.js

@@ -1,7 +1,7 @@
 import { ModelMix } from '../index.js';
 try { process.loadEnvFile(); } catch {}
 
-const mmix = ModelMix.new({ config: { max_history: 10 } }).gpt41nano();
+const mmix = ModelMix.new({ config: { max_history: 10 } }).gpt5nano();
 mmix.setSystem('You are an assistant and today is ' + new Date().toISOString());
 
 // Add web search capability through MCP

package/demo/parallel-strategy.js

@@ -129,7 +129,7 @@ async function genericRLMExample(variableName, variableData, task) {
             }
         } 
     })
-        .gpt41nano()
+        .gpt5nano()
         .setSystem(`You are a Recursive Language Model (RLM) agent.
 
 Data is in the ENVIRONMENT as variables, not in your context. You work programmatically:
@@ -282,7 +282,7 @@ You decide the strategy based on the data and task.`);
                     } 
                 });
                 
-                callMmix.gpt41mini();
+                callMmix.gpt5nano();
                 
                 if (call.system_prompt) {
                     callMmix.setSystem(call.system_prompt);
@@ -376,7 +376,7 @@ You decide the strategy based on the data and task.`);
                 max_history: 10
             } 
         })
-            .gpt41nano();
+            .gpt5nano();
         
         if (system_prompt) {
             recursiveMmix.setSystem(system_prompt);

package/demo/parallel.js

@@ -14,7 +14,7 @@ const mix = new ModelMix({
     }
 })
 
-mix.gpt41nano();
+mix.gpt5nano();
 
 // Function to create a promise that resolves after a random time
 const randomDelay = () => new Promise(resolve => setTimeout(resolve, Math.random() * 2000 + 1000));
@@ -48,4 +48,4 @@ async function runExample() {
 }
 
 // Run the example
-runExample().catch(console.error);
+runExample().catch(console.error);

package/demo/repl-powers.js

@@ -12,9 +12,9 @@ async function replPowersExample() {
     console.log('\n=== JavaScript REPL - Potencias de 2 ===\n');
     const gptArgs = { options: { reasoning_effort: "none", verbosity: null } };
     const mmix = ModelMix.new({ config: { debug: 2, max_history: 10 } })
-        .gpt41nano()
+        .gpt5nano()
         .gpt52(gptArgs)
-        .gemini3flash()
+        .gemini37flash()
         .setSystem('You are a helpful assistant with access to a JavaScript REPL. When you use the REPL and get results, always show them to the user in your response.');
 
     // Variable para capturar el resultado de la herramienta
@@ -74,4 +74,3 @@ try {
 } catch (error) {
     console.error('❌ Error:', error);
 }
-

package/demo/rlm-basic.js

@@ -36,7 +36,7 @@ async function basicRLM() {
     console.log('🎯 Query: "Find top Engineering talents with Python skills"\n');
 
     const mmix = ModelMix.new({ config: { debug: false, max_history: 20 } })
-        .gpt41nano()
+        .gpt5nano()
         .setSystem(`You are an RLM (Recursive Language Model). 
         
 Instead of reading all data at once, you:
@@ -135,7 +135,7 @@ Be strategic: think about what information you need and query only that.`);
 
         // Create a focused recursive call
         const subMmix = ModelMix.new({ config: { debug: false } })
-            .gpt41nano()
+            .gpt5nano()
             .setSystem(`Solve this focused sub-task concisely. Recursion level: ${recursionCount}`);
 
         subMmix.addText(`Context: ${context}\n\nTask: ${task}`);

package/demo/rlm-fast.js

@@ -50,7 +50,7 @@ async function createIvmContext(isolate, contextData, mmixInstance) {
     await jail.set('__mmixCallback', new ivm.Reference(async (system, message, outputJson) => {
         const output = JSON.parse(outputJson);
         const result = await mmixInstance.new()
-            .gpt41nano()
+            .gpt5nano()
             .setSystem(system)
             .addText(message)
             .json(output, output);
@@ -209,8 +209,8 @@ async function demo() {
     // Create base mmix instance for the callbacks
     const model = ModelMix.new({ config: { debug: 2, bottleneck: {} } })
         .gpt52({ options: { reasoning_effort: 'none', verbosity: null } })
-        .gpt41nano()
-        .gemini3flash();
+        .gpt5nano()
+        .gemini37flash();
 
     // Run the IVM task
     const result = await runIvmWithMmix({

package/demo/rlm-simple.js

@@ -105,9 +105,9 @@ async function rlmExample() {
 
     const gptArgs = { options: { reasoning_effort: "none", verbosity: null } };
     const mmix = ModelMix.new({ config: { debug: false, max_history: 15 } })
-        .gpt41nano()
+        .gpt5nano()
         .gpt52(gptArgs)
-        .gemini3flash()
+        .gemini37flash()
         .setSystem(`You are an RLM (Recursive Language Model) agent. 
         
 KEY PRINCIPLE: Instead of processing the entire document directly, you can:
@@ -216,7 +216,7 @@ Current recursion depth: ${recursionDepth}/${maxDepth}`);
 
         // Crear una nueva instancia para la llamada recursiva
         const recursiveMmix = ModelMix.new({ config: { debug: false } })
-            .gpt41nano()
+            .gpt5nano()
             .setSystem(`You are processing a sub-task. Be concise and direct.
 Recursion depth: ${recursionDepth}/${maxDepth}
 ${document_chunk ? 'Document chunk provided.' : 'No document chunk provided.'}`);

package/demo/short.js

@@ -12,7 +12,7 @@ const mmix = await ModelMix.new(setup)
     .sonnet46() // (main model) Anthropic claude-sonnet-4-6
     .gpt56luna() // (fallback 1) OpenAI gpt-5.6-luna
     .gemini37flash() // (fallback 2) Google gemini-3.7-flash
-    .gpt41nano() // (fallback 3) OpenAI gpt-4.1-nano
+    .gpt5nano() // (fallback 3) OpenAI gpt-5-nano
     .grok46() // (fallback 4) Grok grok-4.6
     .qwen35397b() // (fallback 5) OpenRouter qwen/qwen3.5-397b-a17b
     .hermes470b() // (fallback 6) OpenRouter nousresearch/hermes-4-70b

package/demo/stream.js

@@ -1,7 +1,7 @@
 import { ModelMix } from '../index.js';
 try { process.loadEnvFile(); } catch {}
 
-await ModelMix.new().gpt41nano()
+await ModelMix.new().gpt5nano()
     .addImageFromUrl('https://pbs.twimg.com/media/F6-GsjraAAADDGy?format=jpg')
     .addText('describe')
     .stream((data) => { console.log(data.message); });

package/demo/tokens.js

@@ -26,7 +26,7 @@ console.log('-'.repeat(60));
 const providers = [
     { name: 'OpenAI GPT-5-nano', fn: (m) => m.gpt5nano() },
     { name: 'Anthropic Haiku', fn: (m) => m.haiku45() },
-    { name: 'Google Gemini', fn: (m) => m.gemini25flash() }
+    { name: 'Google Gemini', fn: (m) => m.gemini37flash() }
 ];
 
 const prompt = 'Explain quantum computing in one sentence.';

package/demo/verbose.js

@@ -18,7 +18,7 @@ console.log('──────────────────────
 
 await ModelMix
     .new({ config: { verbose: 0 } })
-    .gpt41nano()
+    .gpt5nano()
     .addText(prompt)
     .message();
 
@@ -33,7 +33,7 @@ console.log('──────────────────────
 
 await ModelMix
     .new({ config: { verbose: 1 } })
-    .gpt41nano()
+    .gpt5nano()
     .addText(prompt)
     .message();
 
@@ -48,7 +48,7 @@ console.log('──────────────────────
 
 await ModelMix
     .new({ config: { verbose: 2 } })
-    .gpt41nano()
+    .gpt5nano()
     .addText(prompt)
     .json({ message: 'string' });
 
@@ -63,7 +63,7 @@ console.log('──────────────────────
 
 await ModelMix
     .new({ config: { verbose: 3 } })
-    .gpt41nano()
+    .gpt5nano()
     .addText(prompt)
     .message();
 
@@ -79,7 +79,7 @@ try {
     const resultFallback = await ModelMix
         .new({ config: { verbose: 2 } })
         .attach('fake-model-that-will-fail', new MixOpenAI())
-        .gpt41nano() // This will be the fallback
+        .gpt5nano() // This will be the fallback
         .addText(prompt)
         .message();